Montreal, Quebec, Canada H3C 3J7; and Department ofExperimental Medicine, ..... construct fused to the1.9-kbp mouse enhancer (black bar). Abbre- viations ...
Vol. 14, No. 2
MOLECULAR AND CELLULAR BIOLOGY, Feb. 1994, p. 1084-1094
0270-7306/94/$04.00+0
Specific Expression of the Human CD4 Gene in Mature CD4+ CD8- and Immature CD4+ CD8+ T cells and in Macrophages of Transgenic Mice ZAHER HANNA,l.2* CAROLE SIMARD,1 ANDRE LAPERRIERE,1 AND PAUL JOLICOEUR1'3'4 Laboratory of Molecular Biology, Clinical Research Institute of Montreal, Montreal, Quebec, Canada H2W JR7U; Departments of Medicine2 and Microbiology and Immunology, 3 Universite6 de Montreal, Montreal, Quebec, Canada H3C 3J7; and Department of Experimental Medicine, McGill University, Montreal, Quebec, Canada H3G 1A44 Received 23 August 1993/Accepted 11 November 1993
The CD4 protein plays a critical role in the development and function of the immune system. To gain more insight into the mechanism of expression of the human CD4 gene, we cloned 42.2 kbp of genomic sequences comprising the CD4 gene and its surrounding sequences. Studies with transgenic mice revealed that a 12.6-kbp fragment of the human CD4 gene (comprising 2.6 kbp of 5' sequences upstream of the transcription initiation site, the first two exons and introns, and part of exon 3) contains the sequences required to support the appropriate expression in murine mature CD4+ CD8- T cells and macrophages but not in immature double-positive CD4+ CD8+ T cells. Expression in CD4+ CD8+ T cells was found to require additional regulatory elements present in a T-cell enhancer fragment recently identified for the murine CD4 gene (S. Sawada and D. R. Littman, Mol. Cell. Biol. 11:5506-5515, 1991). These results suggest that expression of CD4 in mature and immature T-cell subsets may be controlled by distinct and independent regulatory elements. Alternatively, specific regulatory elements may control the expression of CD4 at different levels in mature and immature T-cell subsets. Our data also indicate that mouse macrophages contain the regulatory factors necessary to transcribe the human CD4 gene. In addition, the CD4 protein has been found to be expressed on human monocytes, macrophages, and microglial cells of the central nervous system (10, 39). However, this expression appears to be species specific and is not found in murine monocytes or macrophages (7). A CD4-related transcript has been detected in mouse brain (14, 22), but its cellular site of synthesis has not yet been identified. The function of the CD4 protein in these nonlymphoid cells remains unclear. Interestingly, the CD4 molecule has been shown to be the main receptor for human immunodeficiency virus (HIV) (8, 17) in both T cells and macrophages. Following HIV infection, CD4+ lymphoid cells lose surface expression of the CD4 molecule, although the precise mechanisms of this down regulation are not known (15). Thus, establishing the mechanism controlling CD4 expression is important not only because CD4 modulation is a feature of normal T-cell development (19) but also because this molecule is the receptor for HIV. The identification of some of the regulatory elements of the mouse CD4 gene has been reported (34). The CD4 promoter was found, by in vitro transfection studies, to be active through Myb binding sites in mature CD4+ CD8- T-cell lines but not in immature CD4+ CD8+ T-cell lines (34). Recently, the promoter of the human CD4 gene was characterized, and transcription of the gene was found to be activated by Ets proteins (30). In addition, a T-cell-specific enhancer was identified for the murine (33) and human (3) CD4 genes. However, another group was unable to confirm such an enhancer activity in vitro for the murine sequences (34). Recent studies in transgenic mice have confirmed the presence of this T-cell-specific enhancer (3, 16). It was found that the presence of a T-cell-specific enhancer (3, 16) was necessary to achieve an increased level of T-cell-specific transcription. However, conflicting results
The CD4 gene encodes a transmembrane glycoprotein which is a member of the immunoglobulin gene superfamily (for a review, see reference (19). It is coexpressed with the CD8 molecule on immature thymocytes, early in thymic development, giving rise to double-positive CD4+ CD8+ T cells. The regulation of CD4 and CD8 expression is tightly controlled and tied to the maturation of CD4+ CD8+ T cells. Indeed, in the peripheral immune system, the CD4 and CD8 molecules are expressed on mutually exclusive subsets of mature T cells. CD4 is largely expressed on helper T cells that utilize class II major histocompatibility complex (MHC) proteins as restriction elements for antigen recognition, while CD8 is expressed on cytotoxic and suppressor T cells and interacts with targets expressing class I MHC proteins
(19). The CD4 molecule is essential for normal T-cell functions and plays an important role in T-cell development and activation. It is closely associated with the CD3 T-cell receptor (TcR) complex (29) required for antigen recognition by T cells and for the activation of T cells (2, 29). CD4 is also a signal-transducing molecule which is physically associated with the tyrosine kinase p56'ck protein (37). Upon interaction with a specific antigen in the context of MHC class II molecules, CD4+ T lymphocytes produce several lymphokines which induce the proliferation and differentiation of other effector cells of the immune response (19). Therefore, the expression of CD4 on T cells and its interaction with class II MHC proteins is crucial for the development of the cellular immune response. * Corresponding author. Mailing address: Laboratory of Molecular Biology, Clinical Research Institute of Montreal, 110 Pine Ave. West, Montreal, Quebec, Canada H2W 1R7. Phone: (514) 987-5569. Fax: (514) 987-5688.
1084
VOL. 14, 1994
CD4 REGULATORY SEQUENCES IN TRANSGENIC MICE
were reported in studies of transgenic mice containing the human CD4 gene and flanking sequences in the absence of the enhancer. One group failed to demonstrate expression of the transgene in T cells, suggesting that the CD4 promoter itself was insufficient for T-cell-specific expression (16), in contrast to the results obtained in vitro (34). Another group was able to show expression in a small fraction of mouse splenocytes, but most of this expression was in B lymphocytes (3). In the same study, the investigators observed expression of human CD4 in the thymus, but this expression was in non-T cells (3). In independent transgenic mice, Gillepsie et al. (13) also identified sequences of the human CD4 gene able to drive expression of the CD4 gene in CD4+ T lymphocytes of peripheral blood mononuclear cells. However, the CD8 marker was not studied, and expression of the transgene in specific T-cell subsets was not reported. In an effort to understand how the expression of CD4 is controlled during thymocyte development and what factors may be responsible for its complex regulation, we cloned and analyzed the human CD4 gene. Because in vitro transfection studies may not be reliable for identifying the regulatory elements controlling such a complex system, cellspecific expression of the human CD4 promoter/enhancer was studied in transgenic mice. Our results indicate that expression of the human CD4 gene may be controlled by independent regulatory elements active in different subsets of CD4+ T cells. In addition, cellular sequences conferring macrophage-specific expression were identified upstream of the CD4 coding sequences.
MATERIALS AND METHODS Molecular cloning of genomic CD4 sequences. A human genomic library was prepared essentially as described previously (23, 28). Approximately 106 plaques were screened, using the human CD4 cDNA clone (21) as a probe. Eleven positive clones were isolated and mapped by standard single and double digestion with various restriction endonucleases (P. L. Pharmacia, Montreal, Quebec, Canada). The mouse CD4 enhancer was isolated from a mouse phage genomic library by using mouse CD4 cDNA clone (20). Southern blot analysis. DNA analyses by agarose gel electrophoresis and Southern blotting (36) were done on nylon membranes (Amersham). Probes were 32P labeled with mixed oligonucleotides as primers (11). Northern (RNA) blot analysis. RNA was isolated by the method of Chomczynski and Sacchi (6). RNA samples were applied to 1% agarose gels containing formaldehyde as described previously (23). Following electrophoresis, RNAs were blotted onto nylon membranes (Amersham). After UV treatment, blots were hybridized with the 32P-labeled 1.8-kb EcoRl-BamHI human CD4 cDNA fragment (21) at 65°C and washed in conditions previously described (28). RNase protection analysis. RNase protection analysis was performed essentially as described by Melton et al. (25). Genomic fragments were cloned in the vector pGEM-3, and 0.5 Fg of linearized DNA was used for synthesis of 32p_ labeled antisense RNA probe. Two probes were used in this analysis. Probe A consisted of a 0.5-kbp genomic DNA fragment extending upstream from the RsaI site (within exon 3) to the NcoI site upstream to exon 2. Probe B consisted of a 0.6-kbp genomic fragment 10 kbp upstream from exon 2, extending from the XbaI site through exon 1 to the PvuII site. These antisense 32P-labeled RNA probes were hybridized with total RNA (20 to 30 ,g) isolated from CEM and MT-4 CD4+ lymphocytes or from transgenic organs. As a
1085
negative control, an equivalent amount of RNA from nontransgenic mouse spleen was used. RNase-protected fragments were fractionated on 8% denaturing polyacrylamide gels (27). Flow cytometry. The antibodies used in this study, phycoerythrin (PE)-coupled anti-human CD4 antibody Leu3a (Becton Dickinson), fluorescein isothiocyanate (FITC)-coupled antibody anti-murine CD4 GK 1.5), biotin-coupled anti-murine CD8 antibody Ly2, and FITC-coupled antimurine CD3 antibody 145-2C11, were previously described (27). The cell suspensions obtained from lymphoid organs were stained with antibodies as previously described (27). Detection of biotinylated antibodies was facilitated by using PE-avidin (Southern Biotechnology) as a second-step reagent. Cells were analyzed on a FACScan (Becton Dickinson) as described previously (27). Sequencing and computer analysis. Sequencing was performed by the dideoxy-chain termination method of Sanger et al. (32). The DNA sequence was determined by isolation and subcloning of specific restriction fragments into either pGEM-3 or pUC-18 vector essentially as described (1). The entire nucleotide sequence of the promoter was determined from both strands, and all sites used for cloning were crossed in the sequencing. [a-35S]dATP (Amersham) and Sequenase (United States Biochemical) were used in sequencing reactions. Sequence alignments and homology and motifs searches were performed with the Genetics Computer Group software package (9). Construction of transgenes. To construct transgene CD4A/ CD4, the 5' upstream 13-kbp SalI-RsaI fragment of the human CD4 promoter, containing exon 1, intron 1, exon 2, and part of exon 3 up to the RsaI site, was fused to the RsaI site of a 1.7-kbp fragment of the human cDNA sequences derived from plasmid pT4B (21) and a 1.9-kbp SalI-EcoRI simian virus 40 region harboring a polyadenylation site. The cloned fragment to be microinjected was isolated as a 16.6-kbp fragment by EcoRI and partial SalI digestion. Transgene CD4B/CD4 was constructed by deleting a 4.8-kbp SacI 5'-end fragment (containing exon 1 and part of intron 1) from the transgene CD4A/CD4 (see Fig. 2). The fragment to be microinjected was excised as an 11.8-kbp EcoRI fragment. Transgene CD4C/CD4 was constructed by ligating a 1.9-kbp EcoRI-SalI fragment containing the mouse CD4 T-cell-specific enhancer (33) to transgene CD4A/CD4. The transgene DNA used for microinjection was excised as a 18.5-kbp EcoRI fragment. DNAs, purified by preparative agarose gel electrophoresis, were banded on CsCl as previously described (4, 27). Production of transgenic mice. One-cell (C57BL16 x C3H)F2 embryos were collected, microinjected, and transferred into pseudopregnant CD1 females essentially as described previously (4, 27). The presence of the transgene was confirmed by Southern hybridization with tail DNA, using CD4 cDNA as a probe. Five CD4A/CD4 transgenic founders (5566, 7753, 8263, 8277, and 8786) were produced from 69 pups born. Three CD4B/CD4 (Dl, D2, and D9) and four CD4C/CD4 (8918, 8929, 8951, and 8956) transgenic founders were obtained from 11 and 27 pups born, respectively. All founders transmitted the transgene in a Mendelian fashion and appeared phenotypically normal. RESULTS Cloning of the human CD4 gene and its flanking sequences. A human placenta genomic library was screened by using the 1.8-kbp cDNA fragment of the human CD4 gene as a probe
1086 A
HU}I cDNA~ .
HANNA ET AL.
MOL. CELL. BIOL.
OVERLAPPING X CLONES
AX13
1.,
CD4A/CD4
X41
1-6
P
SIBg I,I
X 21
S
X 51
CD4 GENE 5'
l
//
I
S B
HUMAN gDNA
HUMAN cDNA CHIMP cDNA
HUMAN gDNA HUMAN cDNA CHIMP cDNA HUMAN gDNA HUMAN cdNA
B
(sR.I ,Bg
Bg
Bg
Bg
Rs
Si
R
S
2Rs S IR *tzzi=I
11*H= -F-- 31
,. .11
H BH B
H
H H
B
BB
B
R H lKbp
Si Bg
1
I
y
S
S
r U4I4 Ir.r'. s.(LUQ
2Rs
SI
R
S I-I
1 Kpb
,.o
r
CHIMP cDNA
CDBCD CD4B/CD4
45 6789 10
XX RV3 Bg Bg Rv Bg . .I.III ., Y I II S SH H BH - B Ha
CHIMP cDNA
S
Bg
.
H
SI /Bg/ t IV r .
23
I
q
1Kbp
I
24 mer
CCCCACTGGGCTCCTG0TTGCAGAG,'CTCCAAGTCCTCACACAGATACGCC
50
EXON 1 - IEXON 2 TGTTTGAGAAGCAGCGGGCAAGAAAGACGCAAGCCCAGAGGCCCTGCCAT
100
INTRON 1 < , EXON 2 CAG.AA.TGTTTGCT. ACT. .TG.TTGGCATTT ... TC .._ EXON 1 EXON 2 *4 -
..........._
TTCTGCGGGCTCAGGTCCCTACTGGCTCAGGCCCCCGCCTCCCTCAGCAA .....
T .... T.
..... T .... T.
GGCCACAATGAACCGGGGAGTCCCTTTTAGGCACT
150
FIG. 2. Structures of the CD4A/CD4, CD4B/CD4, and CD4C/ CD4 transgenes. The CD4A/CD4 construct contains 13 kbp of human genomic CD4 sequences (including 2.6 kbp 5' to exon 1), exons 1 and 2, and part of exon 3 up to the RsaI site (black boxes). This fragment was fused to the RsaI site of the human CD4 cDNA (open box) and simian virus 40 fragment containing a polyadenylation signal (hatched box). The CD4B/CD4 transgene was constructed by deleting the 5' SacI fragments from the CD4A/CD4 DNA. The CD4C/CD4 construct contains the entire CD4A/CD4 construct fused to the 1.9-kbp mouse enhancer (black bar). Abbreviations are as defined in the legend to Fig. 1.
.. G
G .... 185
...
...
FIG. 1. (A) Structure of the human CD4 gene. A partial restriction map of the gene with exons (black boxes) numbered 1 to 10 and introns (open bar) is shown. The region covered by the four phage clones (X13, X21, X51, and X41) isolated from the human placenta DNA is indicated by arrows at the top. Restriction endonucleases: B, BamHI; Bg, BgII; H, HindIII; Ha, HaeIII; RV, EcoRV; S, SacI; Sl, SalI; X, XbaI. The Sall site belongs to the lambda phage vector. (B) Nucleotide sequence alignment of human genomic DNA (gDNA) around exon 2 and the 5' end of the published sequences of the human (21) and chimpanzee (5) CD4 cDNAs. Dots represent nucleotides identical to those in the chimpanzee cDNA, and dashes represent the absence of nucleotides. The initiation codon ATG is underlined. The 24-mer oligonucleotide used to locate exon 1 is indicated.
(21). Several overlapping phage clones were isolated, and four of them, designated X13, X21, X51, and X41, were selected (Fig. 1A). Restriction analysis and Southern blot hybridization with probes derived from the cDNA revealed that the length of the cloned genomic DNA fragment containing all of the cDNA sequences was 42.2 kbp (Fig. 1A). The DNA inserts of X13, X21, X51, and X41 were assembled into one large fragment (42.2 kbp) and cloned in the cosmid pWE15. Restriction analysis and Southern hybridization established the colinearity of the cosmid clone with the authentic human CD4 gene (data not shown) and suggested that substantial structural alterations did not occur during cloning of the gene. To identify the regplatory sequences controlling this gene, we first sequenced the region around its first translated exon. Alignment of this genomic sequence with the nucleotide sequences of previously defined human (21) and chimpanzee (5) cDNA clones, both of which contain part of the 5' untranslated region (Fig. 1B), suggested that the human CD4 gene has an additional exon. To determine the location of this exon in the human CD4 gene, we synthesized a 24-base oligonucleotide complementary to the 5' end of the chimpanzee cDNA (Fig. 1B) and used it as a probe on the genomic X13 clone. We were able to
locate this exon around the Sacl site and at -10.5 kbp upstream from exon 2 (Fig. 1A). Therefore, in contrast to published results (22), the human CD4 gene was found to comprise 10 rather than 9 exons, interrupted by nine introns of various lengths (-0.13 to 14.2 kbp), similar to the mouse CD4 gene (14). Construction of transgenic mice harboring the human CD4 upstream sequences. To test whether the human CD4 upstream sequences are capable of T-cell-specific transcriptional activation in vivo, as well as to identify their cis-acting regulatory regions, we constructed three transgenes containing upstream sequences of the CD4 gene linked to the human CD4 cDNA coding sequences as a reporter gene (Fig. 2). The first transgene (CD4A/CD4) contained 12.6 kbp of 5' genomic sequences, including 2.6 kbp of 5' upstream sequences, exons 1 and 2, introns 1 and 2, and part of exon 3 up to the RsaI site. These sequences were fused to the RsaI-SalI human CD4 cDNA fragment derived from plasmid pT4B (21) and to the simian virus 40 fragment containing a polyadenylation signal. Transgene CD4B/CD4 was constructed by deleting 5 kbp of upstream genomic sequences from transgene CD4A/CD4 (Fig. 2), thus deleting the promoter, exon 1, and part of intron 1. Transgene CD4C/CD4 was constructed by adding a 1.9-kbp mouse CD4 enhancer to the CD4A/CD4 transgene. This murine T-cell-specific enhancer located 13 kbp upstream of the first exon of the mouse CD4 gene (33) and its human homolog (3) were identified while our work was in progress. In vitro analysis showed that the enhancing activity of the murine fragment was not restricted to CD4+ T cells but was also observed in CD4- CD8+ cell lines (33). However, another group was unable to confirm such an enhancer activity in vitro (34). Therefore, we decided to test the activity of this enhancer by in vivo studies. Five, three, and four transgenic founders were produced with the transgenes CD4A/CD4, CD4B/CD4, and CD4C/CD4 DNA, respectively. Lines were derived from each of these founders. Transgene RNA expression in the CD4A/CD4, CD4B/CD4, and CD4C/CD4 transgenic mice. To determine the expression and tissue distribution of the transgene RNA in these transgenic mice, we used RNase protection and Northern blot
VOL. 14, 1994
I~
CD4 REGULATORY SEQUENCES IN TRANSGENIC MICE
analysis on total RNA isolated from different organs of adult mice. In CD4B/CD4 transgenic mice, low expression was detected by RNase protection with probe A in almost all lymphoid as well as nonlymphoid tissues (data not shown), suggesting the presence of weak promoter activity which is not tissue specific in this transgene and indicating that the sequences conferring tissue-specific transcription to the CD4 gene lie outside the region included in this transgene. A similar analysis was carried out on tissues of the CD4A/CD4 transgenic mice (Fig. 3A). Four founders (5566, 7753, 8263, and 8277) showed similar patterns of expression, while one founder (8286) did not express the transgene. Transgene RNA expression was the highest in the spleen, lymph nodes, and macrophages. Expression was also found, but to a lesser extent, in the thymus, brain, and intestines and was absent in all other organs tested except the liver and lungs. The weak signal in these latter two organs may reflect the presence of T cells and macrophages, which were found to express the transgene (see below). Similar results were obtained with another probe (probe B) (data not shown). These results indicate that this transgene harbors all sequences necessary to direct expression in macrophages and in lymphoid organs. Transgene expression in CD4C/CD4 mice was significantly higher than in CD4A/CD4 mice and could be measured by Northern blot analysis. The levels of transgeneencoded transcripts in mice from founder 8951 were highest in the thymus and found to a lesser extent in the spleen and lymph nodes (Fig. 3B). Transgene expression was negative in all other organs tested but detectable in the brain and lungs. The weak signal in these latter two organs may reflect the presence of macrophages, which were found to express the transgene (see below). We also observed weak expression in the intestines, reflecting expression in resident T cells of Peyer's patches (Fig. 3B, lanes 12 and 13) (14a). Similar results were obtained with two other founders (8929 and 8956). These results showed that the CD4C/CD4 transgene contains the regulatory sequences necessary to direct high expression of human CD4 in lymphoid organs. Expression of the human CD4 cell surface protein in lymphoid cells of CD4A/CD4 and CD4B/CD4 transgenic mice. To measure the levels of expression of the human cell surface CD4 protein on lymphoid cells of transgenic mice, we used flow cytometry with a monoclonal antibody specific to the human CD4 protein. In CD4B/CD4 transgenic mice, no expression of human CD4 was detected in spleen or thymus cells of mice from the three different founder lines (data not shown). In CD4A/CD4 transgenic mice from founder line 5566, the human CD4 protein was expressed in 11.2% + 4.2% (n = 7), 17.7% + 3.6% (n = 8), 28.0% ± 3.3% (n = 5), and 11.8% 1.8% (n = 5) of the total cell population of the thymus, spleen, mesenteric lymph node, and circulating mononuclear cells, respectively (Table 1). CD4A/CD4 mice from founders 7753 and 8277 also expressed the human transgene in their lymph nodes and thymuses but at a lower percentage (Table 1). To determine in which cell populations the human CD4 protein was expressed, we used a two-color immunofluorescence fluorescence-activated cell sorting (FACS) analysis with anti-human CD4 and anti-mouse CD4 or anti-mouse CD8 monoclonal antibodies. In the peripheral lymphoid organs (spleen and lymph nodes) of mice from all three CD4A/CD4 founders, the human CD4 protein was expressed almost exclusively on mouse mature CD4+ CD8- T cells (over 95%) (Fig. 4B; Table 1) and on very few of mature
A
1
2 3 4 5 6 7 8 9 10 11 12 13 14
__ _ w
B Kb
2 3
4.84.33.0 2.31-
t.72.3.1.70.98-
0.9 -
a
1087
4
5
6
7
8
9
-
-147
* -123 * -110
Ir
--74
9
-* *
-67
.
-55
10 11! 12 13 14 15 16 to
-CD4
-GAPDH
FIG. 3. (A) RNase protection analysis of RNA from different organs of CD4A/CD4 transgenic mice (founder 5566). Total RNAs (20 pg) from different organs were used with 32P-labeled probe A. After digestion with RNases, the fragments were run on an 8% polyacrylamide gel and visualized by autoradiography. A 74-base protected fragment constituting part of exon 3 was taken as a diagnostic fragment for the presence of the transgene. The other protected fragments were nonspecific and appeared also in the negative control. Lanes: 1 and 14, markers; 2 to 11, RNAs from lungs, intestine, kidney, liver, mesenteric lymph node, brain, spleen, thymus, muscle, and macrophages, respectively; 12, RNA from a nontransgenic spleen; 13, RNA from human CEM cells as a positive control. The protected fragment (74 bases) is indicated. By using a densitometer, the relative levels of RNA in lymph node (lane 6), spleen (lane 8), and thymus (lane 9) were calculated to be 4.1, 6.3, and 1.0 respectively. (B) Northern blot analysis of human CD4 RNA extracted from different tissues of a CD4C/CD4 transgenic animal (founder 8951). Lanes: 1, RNA marker; 2, RNA from MT-4 cells (positive control); 3 to 16, RNAs from sciatic nerve, salivary gland, lymph nodes, liver, bone marrow, thymus, heart, kidney, lungs, Peyer's patches, intestine, brain, spleen, and muscle, respectively. The probe used for hybridization was the 32P-labeled human CD4 cDNA. GAPDH, glutaraldehyde 3-phosphate dehydrogenase.
1088
HANNA ET AL.
MOL. C-ELL. BIOL.
TABLE 1. Expression of human CD4 in lymphoid organs of CD4A/CD4 and CD4C/CD4 transgenic micea % ± SE (n) Founder hCD4
CD4A/ CD4 5566 28.0 ± 3.3 (5) 7753 11.2 ± 3.2 (4) 8277 23.2 ± 4.8 (6)
Mesenteric lymph node cells that are hCD4 that are hCD4mCD4 mCD8 97.9 ± 1.0 (5) 97.9 ± 0.7 (4) 95.0 ± 2.2 (5)
Thymic cells mCD4 that are hCD4
hCD4
hCD4 that are mCD4
0.5 ± 0.3 (2) 58.4 ± 12.2 (5) 11.2 ± 4.2 (7) 99.6 ± 0.3 (7) 0.7 ± 0.4 (4) 23.3 ± 6.6 (3) 5.5 ± 2.0 (5) 98.7 ± 0.8 (4) 1.6 ± 1.1 (5) 40.5 ± 7.1 (5) 5.6 ± 1.5 (5) 97.4 ± 1.9 (4)
hCD4 that are mCD8
mCD4 that are hCD4
6.0 ± 1.9 (3) 12.3 ± 5.3 (7) 6.9 ± 1.8 (3) 6.6 ± 2.3 (6) 7.4 ± 2.3 (3) 6.7 ± 1.3 (4)
CD4C/ CD4 8951 63.3 ± 15.2 (6) 87.8 ± 6.7 (4) 11.8 ± 0.4 (2) 99.7 ± 0.4 (6) 8956 58.9 ± 4.9 (7) 82.0 ± 8.4 (5) 13.0 ± 3.0 (3) 99.2 ± 1.0 (7) 8929 46.2 ± 3.1 (2) 92.9 ± 2.2 (2) 6.2 ± 5.5 (2) 97.7 ± 1.9 (2)
94.7 ± 3.6 (6) 95.9 ± 1.7 (4) 74.3 ± 9.0 (3) 98.9 ± 1.3 (6) 96.2 ± 2.0 (6) 94.2 ± 3.3 (5) 73.9 ± 4.2 (3) 99.5 ± 0.3 (6) 92.8 ± 0.8 (2) 95.7 ± 0.3 (2) 77.7 ± 0.4 (2) 98.8 ± 0.6 (2)
a Prefixes "h" and "m" denote human and mouse, respectively.
CD4- CD8+ T cells (less than 1.6%) (Fig. 4D; Table 1). Expression of human CD4 could not be detected on B220+ B cells (data not shown). In the thymus, the CD4 molecule is normally expressed in both mature CD4 CD8- and immature CD4+ CD8+ T cells. The two-color immunofluorescence analysis revealed that human CD4 was expressed almost exclusively on mouse CD4+T cells (over 97%) (Fig. 4H; Table 1) and on very few mouse CD8+ T cells (Fig. 4J; Table 1). To examine on which thymic CD4+ T-cell subsets the human CD4 was expressed, we carried out a three-color FACS analysis on thymocytes of the transgenic animals, using monoclonal anti-human CD4, anti-mouse CD4, and anti-mouse CD8 antibodies. As expected, human CD4 was found to be expressed almost exclusively on mature CD4+ CD8- T cells (94.0%) (11% of 11.7%) but not on immature double-positive CD4+ CD8+ thymic T cells (Fig. 4K; Table 1). Similar findings were obtained with mice from two other founder lines of these CD4A/CD4 transgenic mice (data not
shown). Human CD4 was expressed at low levels (compared with levels found in human CEM cells in culture) on the surface of mature CD4+CD8- T cells of thymus, spleen, or mesenteric lymph nodes (Fig. 5). In addition, only a fraction of these cells (a mean of 23 to 58% of total mature spleen, mesenteric lymph node, or thymus CD4 CD8- T cells, depending on the founder line analyzed) expressed human CD4 (Fig. 4B, E, and K; Table 1). However, after stimulation of splenocytes in vitro with concanavalin A, a T-cell-specific mitogen, the percentage of mouse CD4+ CD8- T cells expressing the human CD4 molecule increased significantly to 86%, as shown in Fig. 4E and F for a representative mouse of the founder line 5566. Expression of the transgene was also evaluated on thymocytes and splenocytes of embryos and young mice at different time points after birth. The human CD4 molecule was not expressed in these cells of day 19 embryos or on lymphoid cells of the neonates (data not shown). However, the percentage of mouse T cells expressing human CD4 progressively increased with age both in the thymus and in the spleen (data not shown). Together, these data clearly show that the CD4A/CD4 transgenic mice express human CD4 in mature CD4 CD8T cells but not in immature CD4+ CD8+ T cells, which represent approximately 80% of thymic T cells. These results suggest that the CD4A/CD4 transgene contains the cis-acting elements required to direct expression in CD4+
mature T cells but is lacking those elements necessary for proper expression in immature double-positive T cells. These data also imply that expression of CD4 in mature and immature T cells is controlled by different regulatory elements or alternatively that some regulatory elements control the expression of CD4 at different levels in mature and immature T-cell subsets. Expression of the human CD4 cell surface protein in lymphoid cells of CD4C/CD4 transgenic mice. To examine the effect of the mouse T-cell-specific enhancer on expression of the human CD4 gene, FACS analyses of cells from the lymphoid organs of mice from four different founders of the CD4C/CD4 transgenic mice were performed. A significant increase in the percentage of human CD4-expressing lymphoid cells of the spleen, compared with that of the CD4A/ CD4 transgenic mice, was found (Fig. 5). In mice from founder 8951, human CD4 was expressed in 63.3% + 15.2% (n = 6) and 94.7% ± 3.6% (n = 6) of the total lymphoid cells of the mesenteric lymph nodes and thymus, respectively (Table 1). Similar results were obtained with the two other CD4C/CD4 founders (8929 and 8956) (Table 1). The fourth founder (8918) expressed CD4 on fewer cells and was not studied further. The intensity of the cell surface expression of human CD4 was also increased significantly in these mice and was comparable to the expression on CEM cells (Fig. 5). To determine the distribution of human CD4 expression in the lymphoid cell subsets, a two-color FACS analysis was conducted on cells from the mesenteric lymph nodes and thymuses as described above. The analysis indicated that in the mesenteric lymph nodes of all three founders, human CD4 was expressed on the majority (over 87%; founder 8951) of mature CD4+ CD8- T cells (Fig. 6B; Table 1). In contrast to the CD4A/CD4 transgenic mice, about 11.8% of the mesenteric lymph node cells expressing human CD4 were mature CD4- CD8+ T cells (Fig. 6D; Table 1). Although almost all (over 97.7%; Table 1) of the mouse CD4+ cells coexpressed human CD4, only a subset (46.8%) of the mouse CD8+ T cells coexpressed it, as shown for a representative mouse in Fig. 6D. In the thymus, human CD4 was expressed on 95.9% of mouse CD4+ and 74.3% of mouse CD8+ T cells of founder 8951 (Fig. 6I; Table 1). To determine which subsets of thymic CD4+ or CD8+ T cells expressed the transgene, a three-color FACS analysis was performed. This analysis revealed that human CD4 was expressed at high levels on immature CD4+ CD8+ and mature CD4+CD8- T cells. On
CD4 REGULATORY SEQUENCES IN TRANSGENIC MICE
VOL. 14, 1994
SPLEEN
MESENTERIC LN
1089
THYMUS
Tg9
Tgi
CD4A/CD4
hCD4
Q1 D 3Q6
C 0.1
CD4C/CD4
.3
hCD4
13.3
10.31
mCD8
CEM
SPLEEN
Tg+
Tg E
9.8
31.6
^1.4~~ Xt6l-W1'S 5
5.,1
1.2
18.6 F
FIG. 5. Comparative levels of expression of human CD4 on human CEM cells and on splenocytes and thymocytes of CD4A/ CD4 and CD4C/CD4 transgenic mice. Cells were harvested from the spleens and thymuses of transgenic CD4A/CD4 (founder 5566) (A and B) and CD4C/CD4 (founder 8951) (C and D) mice or were obtained from the established human CEM T-cell line (E). Cells were stained in a direct assay with anti-human CD4 antibody Leu3A-PE. The vertical bar represents the lower limit of positivity determined by direct labeling of the splenocytes or thymocytes of a nontransgenic littermate, using the same antibody.
hCD4
mCD4
THYMUS GQ1
0.1
11.3
hCD4
mCD4
I
0
mCD4+ CELLS --
Q.1j 106
Q6
K 11o
0'7
hCD4
79.61
81.0
79
80.41
mCD8
FIG. 4. Detection of human CD4 on lymphocytes of CD4A/CD4 transgenic mice. Cells were harvested from mesenteric lymph nodes (LN), spleens, or thymuses of transgenic (Tg+) CD4A/CD4 (founder 5566) mice or nontransgenic (Tg-) littermates and were processed as described in Materials and Methods. In panels A to D, mesenteric lymph node cells of a nontransgenic (A and C) or transgenic (B and D) mouse were doubly labeled with anti-human CD4 monoclonal antibody Leu3A-PE and anti-mouse CD4 (mCD4) monoclonal antibody GK 1.5-FITC (A and B) or with anti-mouse CD8 monoclonal antibody Ly2-FITC (C and D). In panels E and F, spleen cells from a transgenic mouse, before (E) or after (F) stimulation in vitro by concanavalin A for 2 days, were doubly labeled with anti-human CD4 and anti-mouse CD4 antibodies. In panels G to K, thymocytes of nontransgenic (G and I) or transgenic (H, J, and K) mice were labeled with anti-human CD4 monoclonal antibody Leu3A-PE and anti-mouse CD4 monoclonal antibody GK 1.5-FITC (G and H) or with anti-mouse CD8 monoclonal antibody Ly2-biotin-avidin Cy (I and J). In panel K, the cells represented in panel J were labeled in a three-color analysis with monoclonal antibodies anti-human CD4 Leu3A-PE, anti-mouse CD4 GK 1.5FITC, and anti-mouse CD8 Ly2-biotin-avidin Cy. In this panel, only GK 1.5-positive (CD4+) T cells are represented. Note that most cells expressing the human CD4 molecule are mature CD4+ CD8- T cells rather than immature CD4+ CD8+ T cells.
gated mouse CD4+ thymic cells (Fig. 6J), human CD4 was expressed on 99.9% (75.6% of 75.7%) of the CD4+CD8+ T cells and on 99.6% (24.2% of 24.3%) of the CD4+ CD8- T cells. In addition, on gated mouse CD8+ thymic cells (Fig. 6G), human CD4 was expressed on 51.5% (1.7% of 3.3%) of the CD4- CD8+ thymic T cells. These results indicate that the mouse T-cell enhancer present in this transgene not only enhanced the levels of expression of the human CD4 surface protein in mature CD4+ subsets but also facilitated its expression on immature double-positive CD4+ CD8+, as well as a subset of CD4CD8+ T cells. There was no expression of human CD4 on CD4- CD8- thymocytes, indicating that the CD4C/CD4 transgene is capable of driving CD4 expression at the correct stage of thymocyte development. Expression on mature CD4- CD8+ cells suggests that the CD4C/CD4 construct is still missing negative regulatory element(s) required for totally silencing expression of the CD4 gene in this subset. Expression of the human CD4 cell surface protein in macrophages of CD4A/CD4 and CD4C/CD4 transgenic mice. In humans, the CD4 protein is expressed not only on lymphoid T cells (19) but also on macrophages (39). However, in contrast to human CD4, mouse CD4 is not expressed on macrophages (7). It was therefore of interest to monitor expression of the human CD4 protein in macrophages of the CD4A/CD4 and CD4C/CD4 transgenic mice. Mouse peritoneal macrophages were collected and doubly stained with anti-human CD4 and macrophage-specific Mac-1 antibodies. Over 33% of the macrophages from the freshly isolated peritoneal cells from founder 5566 of the CD4AICD4 transgenic mice expressed human CD4 (Fig. 7B). However, after plating on tissue culture dishes and incubation in vitro for 24 h, nearly 100% of adhering macrophages of all tested mice expressed human CD4 (Fig. 7D; Table 2). Similar findings were obtained with mice from founder 8277 of the CD4A/
1090
HANNA ET AL.
QW
A
16.6
08
A 57.8
hCD4
30.6.:
52o0 t- 25.5l mCD4
C
O.t
0,1 D 68.3
01
0.2
0.2 B
..
hCD4
t4i9
- ~ . s u~- . ~ . Tg-
MESENTERIC LN TgTgq
: .-...:. .... : ..
MOL. CELL. BIOL. Tq+ 5.8 ...
2t5 ...................
- :6Q0117 .6 ........ : .. . .:'.. . -1 MAC
...:.:..:..;.... .. 5
8.1
hCD4 MAC-i
D13.3 82 mCD8
THYMUS Tg-
Tgt
~-mCD8+ CELLS--\
hCD4
_mCD4+ CELLS-\ SH
0.1
1.Q
76.9 j
24.2
75.6
0.1
0.1
hCD4 ;^.:' 8.9^
81.0
1.0
1.1
mCD8 FIG. 6. Detection of human CD4 on lymphocytes of CD4C/CD4 transgenic mice. Cells were harvested from mesenteric lymph nodes (LN) or thymuses of transgenic (Tg+) CD4C/CD4 mice (founder 8951) or nontransgenic (Tg-) littermates and were processed as described in Materials and Methods. In panels A to D, mesenteric lymph node cells of a nontransgenic (A and C) or transgenic (B and D) mouse were doubly labeled with anti-human CD4 monoclonal antibody Leu3A-PE and anti-mouse CD4 (mCD4) monoclonal antibody GK 1.5-FITC (A and B) or with anti-mouse CD8 monoclonal antibody Ly2-FITC (C and D). In panels E, F, H, and I, thymocytes of a nontransgenic (E and H) or transgenic (F and I) mouse were doubly labeled with anti-human CD4 monoclonal antibody Leu3A-PE and anti-mouse CD4 monoclonal antibody GK 1.5-FITC (E and F) or with anti-mouse CD8 monoclonal antibody Ly2-biotinavidin Cy (H and I). In panels G and J, the cells represented in panel F and I were labeled in a three-color analysis with monoclonal antibodies anti-human CD4 Leu3A-PE, anti-mouse CD4 GK 1.5FITC, and anti-mouse CD8 Ly2-biotin-avidin CY. In panel G, only Ly2-positive (CD8+) T cells are represented; in panel J, only GK 1.5-positive (CD4+) T cells are represented. Note that most mature CD4+ CD8- and immature CD4+ CD8+ T cells are expressing the human CD4 molecule.
CD4 transgenic mice (Table 2). The same analysis was performed on macrophages of mice from CD4C/CD4 transgenic lines, and macrophage expression of cell surface CD4 was detected in two of three founders tested (Fig. 7D; Table 2) at levels comparable to those found in CD4AICD4 trans-
hCD4
FIG. 7. Detection of human CD4 on macrophages of CD4A/CD4 and CD4C/CD4 transgenic mice. Transgenic CD4A/CD4 (founder 5566) and CD4C/CD4 (founder 8951) mice and a normal littermate were injected intraperitoneally with 1 ml of mineral oil 5 days prior to FACS analysis. Peritoneal cells were harvested and processed as described in Materials and Methods. A portion of these cells was directly labeled (A and B) or plated on tissue culture dishes overnight at 37°C in a CO2 atmosphere (C and D). In panels A and B, cells were doubly stained with anti-Mac-1 monoclonal antibodyrat anti-goat FITC-coupled antibody (in an indirect assay) and anti-human CD4 (hCD4) monoclonal antibody Leu3A-PE. Panel A represents a nontransgenic (Tg-) mouse, and panel B represents a transgenic (Tg+) CD4A/CD4 mouse. In panels C and D, the adherent macrophages were detached from petri dishes and labeled with anti-Mac-1 antibody-rat anti-goat FITC-coupled antibody (C) or with anti-human CD4 monoclonal antibody Leu3A-PE (D). All cytometric analyses were done with a FACScan (Becton Dickinson) apparatus. Symbols: solid lines, nontransgenic mouse; dotted lines, transgenic CD4A/CD4 mouse; dashed lines, transgenic CD4C/CD4 mouse. The slightly higher fluorescence intensity levels seen in the CD4C/CD4 mouse (compared with the CD4A/CD4 mouse) shown in panel D was not seen in every animal analyzed. For most mice studied, these values were comparable between lines CD4A/CD4 and CDC/CD4.
genic mice. This result indicates that the T-cell-specific enhancer has little if any role in expression of human CD4 in macrophages. Together, these results show that the 5' upstream fragment of the human CD4 gene present in the CD4AICD4 TABLE 2. Detection of human CD4 at the cell surface of plated peritoneal macrophages of CD4A/CD4 and CD4C/CD4 transgenic mice Founder
No. of mice expressing human CD4/ no.
tested'
CD4A/CD4 5566 .7/7 7753 .1/5 8277 .3/3 CD4C/CD4 8951 .4/4 8956 .4/4 8929 .0/3 a Assay done on plated macrophages as described in Materials and Methods. In each positive mouse, all cells expressed human CD4 as shown in Fig. 7D.
VOL. 14, 1994
CD4 REGULATORY SEQUENCES IN TRANSGENIC MICE
transgene contains the cis-acting sequences necessary to specifically direct expression of a reporter gene in macrophages. The expression of human CD4 protein in transgenic mice does not alter the distribution of mouse CD4 and CD8 T cells. Expression of the human CD4 molecule in lymphoid tissues of transgenic mice does not seem to affect the distribution of mouse mature T cells in lymphoid organs (27). To determine whether this was the case in the CD4A/CD4 and CD4C/CD4 transgenic mice, we carried out a FACS analysis on mouse CD4 and CD8 T-cell populations. The sizes of the thymus, spleen, and other lymphoid organs were not significantly different in CD4 transgenic mice compared with normal animals (data not shown). In addition, the patterns of expression of mouse CD4 and CD8 were not significantly altered in these transgenic mice relative to their normal littermates except for a possible modest increase of peripheral CD8+ T cells of the CD4C/CD4 transgenic animal (data not shown). Therefore, the presence of the human CD4 molecule does not appear to significantly affect the distribution of the lymphoid cells in these transgenic mice. Identification of potential regulatory sequences around exon 1 of the human CD4 gene. To identify potential regulatory sequences within the human CD4 gene, we sequenced the region surrounding exon 1 (Fig. 8A). A computer search analysis revealed the absence of CAAT or TATA boxbinding sites and other initiator sequences identified recently (24, 35) but identified a remarkable array of elements and motifs, some of which are characteristic of eukaryotic promoters and regulatory regions found in other genes. For example, the consensus sequence (5'[T/C]AACGG3') which binds to Myb transcription factor was recently found to reside in several locations within the 5' flanking sequences of the mouse CD4 gene and has been postulated to have role in regulation of this gene (34). The sequences also display several unique regions of direct repeats. Comparative analysis of these human sequences with the homologous mouse sequences revealed six highly conserved regions, indicated by blocks A through F in Fig. 8B. The conserved locations of these sequences in both species suggest a functional role. For example, block E contains the sequence 5'CTTCCTG3', described as an Ets-binding element (also known as the PEA3 binding site) and known to interact with the lymphoid-specific transcription factor Ets-1 (18). However, the locations of many other motifs are not conserved in the human CD4 promoter as in the mouse promoter. For example, the three potential Myb binding sites were found at different locations (underlined in the human promoter) (Fig. 8). It remains to be determined which of these specific motifs play a role in regulation of human CD4 gene expression and how they operate. Recently, in vitro studies have shown that transcription of the human CD4 gene is activated by Ets proteins (30).
DISCUSSION In this work, we used transgenic mice to study the regulatory elements controlling expression of the human CD4 gene. In humans, the CD4 gene is expressed in macrophages, in CD4+ CD8+ immature T cells, and in CD4+ CD8- mature T cells. We were able to map some of the elements controlling this cell-specific expression of CD4 within a 12.6-kbp region of the human CD4 gene comprising 2.6 kbp 5' sequences upstream of the transcription initiation site, the first two exons and introns, and part of exon 3. As in human cells, the CD4A/CD4 transgenic mice were able to
1091
A
-465
MYB
CCCAGAGCAAAGGAGCTGTTTGTGGGCTTACCACTGCTGTTCCCATATGCCCCCATGC
-405
CTCCCACTTCTTTCCCCACAGCCTGGTCAGACATGGCACTACCACTAATGGAATCTTTCT
-345
TGCCATCTTTTTCTTGCCGC1TAAQGTGGCAGTGACACTTTGACTCCTGATTTAAGCCT . . .MYB
M.YB
-285 GATTCTGCT
-225
TCCTTGACTTTGGCATTTTCACTTTGACATGTTCCCTGAGAG
PvuII. CCTGGGGGGTGGGGAACCAGCTCCAGCTGGTGACGTTTGGGGCCGGCCCAGGCCTAGGGT ETS . -.-' EXON 1 GTGGAGGAGCCTTGCCATCGGG-QCT GTCTCTCTTCATTTAAGCACGACTCTGCAGAA .
-165
Sacl
-105
GGAACAAAGCACCCTCCCCACTGGGCTCCTGGTTGCAGAGCTCCAAGTCCTCACACAGAT
-45
ACGCCTGTTTGAGAAGCAGCGGGCAAGAAAGACGCAAGCCCAGA4kTAAGGTGGTCAGAC ETS.
16
TCGGCTTCTCCCCGGAGCTGAGAGGGAGGGGAACGTGGGGCAGATGCACAGGAATGTG
76
CTCTGCCCA=TCTGCCCACAGCTCTGGCCACCTTCTCTTGCATT
MYB
136
GGAACTG GTCATGAGCATGCGATTTCCCACTGGAACTGTGAGCTTCCAGAGGTCAGAGACTGTGCTA
196 GACTCCTCTCTGCAGCCCCAGCGTGCACAGCTCAGTGTCCAGAGCAATGGGTGCTCCTTA 256
GAGGAGTAGTGACCTAAATAGCAAGATCAGAGAGGGAGTGAAGACTGGAGACTATCCCAG
316
GCTGGGGAAGGCGTGGAAGGCAACTAGTCGTGGGCAGTGGAGGGGAGAAGACTGGAAGAG GGGAAAAGAGGAGAAAAAGAGTGAAGAAGGGGAGGTTAAAAATTAGAAAGAATAAATAAA
376
436 TATAAGGTGGGAGGAAACCTATAAAAAAGAAATGATGAGTAAAAACGATAAAAACAAGAA
XbaI
496 AAAGAGTAAAAGAAGTGGTATCTAGTCTAGA
B MOUSE
B
-A
HUMAN -302 ACTCCTGATTTAAGC lrATTCTG
262 ATTCCAGACTCCAGCTUATTC -252
AGATTA
GCTGC
TlsTCACTPTGACATGTTCCCTGAGAGCCTGGGGGGTGGGGAACCAGCTC
312 -202
CTAcTTTeqTTTrEGGCA
TmTATTCAACTTCCCCAACAACTGGGGGTGG"GAGGAGGGACTC
CAGCTGGTGACGTT
362 C
T
.E
CCGG
-152 GCCATCG GGCTTCCTe TCTCTC UT
400 ACGACCAGGCTTCC
CAGGCCTAGGGTG
AGGAGC
I 1111111111 IIII.11 1111111 CGTCCGCCGTC CClIC . .xon 1 . ]ACACGACTCTGCAGAAGGA
GAACATCTGTGAAGGCAAA TCTTPCm
L0
-
SacI Exon 1 III 11 1 11 1111 11 III 11111 11 GCAAGACTCTCTTCTTCACTAGGTACCTGTTTGCAAAG XhoI
-102 ACAAAGCAC CCTCCCCACTGGGCTCCTGGTTGCAGAG 448
FIG. 8. (A) Nucleotide sequence of the promoter of the human CD4 gene. The sequence upstream of exon 1 is shown. In addition, the sequences of exon 1 and part of intron 1 are presented. The Myb and Ets motifs found in this sequence are underlined. Symbols: r , the major transcription start point localized by RNase protection analysis (30; unpublished data); 1. , splice donor site. Reference position 1 is identified as the last nucleotide in exon 1 of the CD4 gene. (B) Conserved sequences in the 5' flanking region of the promoter of the CD4 gene. The 240 bp of the human and mouse sequences when aligned showed six conserved regions, designated blocks A through F. The previously identified Myb motifs are underlined in the mouse sequence. The symbols r+ and L, show the transcription start points of the human (30; unpublished data) and mouse (34) genes, respectively.
specifically express the reporter CD4 gene in macrophages and in mature CD4+ CD8- T cells, indicating that the 5' upstream region of the human CD4 gene, present in the CD4A/CD4 transgene, contains sufficient cis-regulatory elements to drive tissue-specific expression in mouse macrophages and mature T cells. Interestingly, Sands and NikolicZugic (31) have recently identified several possible transcriptional control elements, defined as DNase I-hypersensitive (DH) sites, within the murine CD4 gene. Seven of these DH sites were found to be T cell specific, and five of them were located within the 12-kb region around the first intron of the murine CD4 gene. The analogous human region was used in our transgene CD4A/CD4 and probably contains similar T-cell-specific DH sites. The exclusive expression of human CD4 in mature CD4+ CD8- T cells of the CD4A/CD4 transgenic mice is also in agreement with the previously
1092
HANNA ET AL.
reported in vitro analysis of the mouse CD4 promoter (34). Sui et al. (34) found that a 172-bp fragment of the mouse CD4 promoter directed the expression of a luciferase reporter exclusively in mature CD4+ CD8- T-cell lines, not in mature CD4- CD8+ or immature CD4+ CD8+ T-cell lines. This 172-bp fragment contains one of the T-cell-specific DH sites identified by Sands and Nikolic-Zugic in CD4+ CD8- cell lines but not in CD4- CD8+ cell lines (31). The expression of human CD4 in macrophages is of particular interest since the mouse CD4 gene itself is not expressed in macrophages. Therefore, our data indicate that mouse macrophages contain all factors necessary for transcribing the human CD4 gene. Our results also suggest that the endogenous mouse CD4 gene fails to be expressed in mouse macrophages because it either lacks the required macrophage-specific regulatory elements or contains suppressor sequences inhibiting expression. The locations of the regulatory sequences allowing expression in mature CD4+ CD8- T cells and in macrophages are not precisely known and could be anywhere within the 12.6-kbp fragment used in front of the reporter gene. Our data do not allow us to localize them with more precision. However, the in vitro findings of Siu et al. (34) indicate that they may reside within the 2.6 kbp of sequences upstream of the first exon. We have sequenced part of this region and found that it is highly homologous to the mouse 172-bp sequences previously identified (34) and harbors a remarkable number of motifs previously shown to function as regulatory elements in other genes (12, 26). A more refined analysis of this region will be required to identify which individual elements are required for tissue-specific expression in CD4+ CD8- T cells and in macrophages. Although a 12.6-kbp fragment of the CD4 gene was sufficient to express the reporter CD4 cDNA gene in macrophages and in mature CD4+ CD8- T cells, it was not sufficient to support its expression in immature CD4+ CD8+ T cells, indicating that regulatory elements specific for CD4+ CD8+ T cells are still missing in the CD4A/CD4 construct. This result also suggests that for control of CD4 gene expression, these immature CD4+ CD8+ T cells use a different set of regulatory elements than mature CD4+ CD8T cells do, as previously suggested (3). This conclusion was confirmed by the results obtained with the CD4C/CD4 transgenic mice, which expressed their transgene not only in mature CD4+ CD8- T cells but also in immature CD4+ CD8+ T cells. The transgene carried by these mice contains an additional 1.9-kbp mouse fragment harboring a T-cellspecific enhancer identified by Sawada and Littman (33). The different control of human CD4 gene expression in mature and immature CD4+ T cells observed in our transgenic mice confirmed earlier work done in vitro by transient transfection techniques (34) and in transgenic mice (3). This finding was also corroborated by our previous study of a mouse mammary tumor virus (MMTV) promoter (MMTVD) in transgenic mice (27). We found that the MMTVD promoter was able to drive expression of surrogate genes specifically in immature CD4+ CD8+ T cells but not in mature CD4+ CD8- T cells. Together, these results suggest that the regulatory elements present in the MMTVD long terminal repeat or around the CD4 gene are recognized differently in mature and immature T cells, as seems the case in our CD4A/CD4 and CD4C/CD4 transgenic mice. Expression of the human CD4 protein detected in murine macrophages of CD4C/CD4 transgenic mice was at about the same levels as expression found in macrophages of CD4A/ CD4 transgenic mice, suggesting that the T-cell-specific
MOL. CELL. BIOL.
enhancer has little if any effect on levels of expression in macrophages. However, levels of human CD4 expression in murine mature CD4+ CD8- T cells of the CD4C/CD4 transgenic mice were much higher than those in the CD4AI CD4 transgenic animals, approaching the levels found in human CD4+ T cells, confirming the enhancing activity of this 1.9-kbp fragment (33). The expression of CD4 in immature CD4+ CD8+ T cells suggest that this 1.9-kbp mouse enhancer fragment harbors cis-acting elements required for activating human CD4 expression in this T-cell subset. However, the CD4C/CD4 transgene was also expressed inappropriately in CD4- CD8+T cells, indicating that this construct is still lacking the cis-acting element required to silence expression of the human CD4 transgene in these mature CD8+ T cells emerging from their progenitors, the CD4+ CD8+ T cells. Such control is not without precedent, since a silencer element outside the TcR-ax enhancer was shown to inactivate the enhancer effect in TcR-yb T cells and in non-T cells but not in TcR-aot T cells (38). Therefore, additional cis-acting elements may be required to inactivate the expression of CD4 in CD4- CD8+ T-cell subsets. While this work was in progress, three other groups (3, 13, 16) attempted to achieve the appropriate expression of human CD4 in transgenic mice by using CD4 genomic constructs. Gillespie et al. (13) used a construct very similar to our CD4C/CD4 transgene and reported expression in the CD4+ subsets of peripheral blood mononuclear cells. Our results seem to confirm their data. However, since no other cell surface marker was used and since a detailed FACS analysis of the cells expressing human CD4 in the thymus was not carried out, it is difficult to further compare their results with ours. On the other hand, it appears that our data are significantly different from the recent results of the other two groups (3, 16), who used genomic human CD4 transgenes containing all of the exons. In contrast to the CD4+ T-cell-specific expression observed in our CD4A/CD4 transgenic mice, both groups (3, 16) failed to demonstrate any expression of the human CD4 on mouse CD4+ thymocytes of their mice harboring a transgene lacking the CD4 enhancer. One group (3) observed expression in the thymus of these mice, but it was in non-T cells. With respect to the peripheral tissues of the same mice harboring the enhancerless transgene, the results of the two groups diverged. Killeen et al. (16) found no expression of human CD4 in spleen T cells, while Blum et al. (3) obtained a low level of expression in the spleen (4% of CD4+ splenocytes), but most of the human CD4 expression was in B cells. In contrast to these results, we have observed, with the enhancerless CD4A/CD4 transgene, significantly higher levels of expression of human CD4 (in 11 and 18% of the T-cell populations of the thymus and spleen, respectively), and we have not been able to detect expression of human CD4 in B cells of any of our transgenic mice. The basis for these discrepancies is not clear but may be related to the exclusive use of genomic sequences, and specifically the 3'-end CD4 sequences, by these groups. The addition of the human T-cellspecific enhancer to the human genomic CD4 transgene constructed by Blum et al. (3) significantly increased the expression of CD4 in CD4+ CD8- and CD4+ CD8+ T cells and in a small population of CD4- CD8+ thymocytes and led to an increased number of spleen CD4+ CD8- T cells coexpressing human CD4 (11% versus 4% in the enhancerless transgene). These results are in agreement with our data for the CD4C/CD4 trangenic mice, although the enhancing effect in the CD4+ T splenocytes was smaller than the effect that we observed. Similarly, Killeen et al. (16) found that
VOL. 14, 1994
CD4 REGULATORY SEQUENCES IN TRANSGENIC MICE
addition of the murine T-cell-specific enhancer to the human genomic CD4 construct led to an increased expression of CD4 specifically in peripheral CD4+T cells, as we have also observed. However, these investigators (16) did not observe expression of human CD4 in mature CD8+ T cells, as we did in CD4C/CD4 transgenic mice. Together, our results and those of others (3, 13, 16) indicate that expression of the CD4 gene is controlled very similarly in mouse and human cells. This complex expression system in different cell types seems to be controlled by several regulatory elements which appear to be distinct, as previously suggested (3). Some of these elements control expression in mature CD4+ CD8- T cells, others (such as the T-cell-specific enhancer) allow expression in immature CD4+ CD8+ T cells, some others direct expression in macrophages, and finally others seem to be involved in the down regulation of the CD4 gene in CD4- CD8'cytotoxic T cells. Alternatively, one can hypothesize that the same regulatory elements of the human CD4 gene control expression at different levels in different subsets of T cells. Some of the conflicting results obtained by different groups may indicate that the context in which these various elements are assembled may be of great importance. A more refined analysis leading to the identification of each of these elements should explain some of these discrepancies. Finally, these transgenic mice, expressing human CD4 in a tissue-specific manner, may be instrumental in understanding some aspects of HIV pathogenesis, since CD4 appears to be the main receptor for HIV. ACKNOWLEDGMENTS This work was supported by grants from the National Health Research and Development Program to Z.H. and from the Medical Research Council of Canada and the National Cancer Institute of Canada to P.J. C.S. was supported by the Health of Animal Laboratory Division, Agriculture Canada, Sackville, New Brunswick, Canada. We are grateful to R. Axel and Dan R. Littman for providing the human and mouse CD4 cDNA clones, respectively. We thank Ginette Masse, Michel Ste-Marie, and Mario Robert for excellent technical assistance and Claude Cantin for valuable help in cytofluorometric analysis. We thank Marie Bernier for typing the manuscript. REFERENCES 1. Aziz, D. C., Z. Hanna, and P. Jolicoeur. 1989. Severe immunodeficiency disease induced by a defective murine leukaemia virus. Nature (London) 338:505-508. 2. Blue, M. L., D. A. Holfer, J. F. Le,vine, H. Levine, K. A. Craig, J. Breitmayer, and S. F. Schlossman. 1988. Regulation of T cell clone function via CD4 and CD8 molecules: anti-CD4 can mediate two distinct inhibitory activities. J. Immunol. 140:376. 3. Blum, M. D., G. T. Wong, K. M. Higgins, M. J. Sunshine, and E. Lacy. 1993. Reconstitution of subclass-specific expression of CD4 in thymocytes and peripheral T cells of transgenic mice: identification of a human CD4 enhancer. J. Exp. Med. 177:13431358. 4. Bouchard, L., L. Lamarre, P. J. Tremblay, and P. Jolicoeur. 1989. Stochastic appearance of mammary tumors in transgenic mice carrying the MMTV/c-neu oncogene. Cell 57:931-936. 5. Camerini, D., and B. Seed. 1990. A CD4 domain important for HIV-mediated syncytium formation lies outside the virus binding site. Cell 60:747-754. 6. Chomczynski, P., and N. Sacchi. 1987. Single-step method of RNA isolation by acid guanidinium thiocyanate-phenol-chloroform extraction. Anal. Biochem. 162:156-159. 7. Crocker, P. R., W. A. Jefferies, S. J. Clark, L. P. Chung, and S.
1093
Gordon. 1987. Species heterogeneity in macrophage expression of the CD4 antigen. J. Exp. Med. 166:613-618. 8. Dalgleish, A. G., P. C. L. Beverley, P. R. Clapham, D. H. Crawford, M. F. Greaves, and R. A. Weiss. 1984. The CD4 (T4) antigen is an essential component of the receptor for the AIDS retrovirus. Nature (London) 312:763-767. 9. Devereux, J., P. Haeberli, and 0. Smithies. 1984. A comprehensive set of sequence analysis programs for the VAX. Nucleic Acids Res. 12:387-395. 10. Dewhurst, S., M. Stevenson, and D. J. Volsky. 1987. Expression of the T4 molecule (AIDS virus receptor) by human brainderived cells. FEBS Lett. 212:133-137. 11. Feinberg, A. P., and B. Vogelstein. 1983. A technique for radiolabeling DNA restriction endonuclease fragments to high specific activity. Anal. Biochem. 132:6-13. 12. Ghosh, D. 1990. A relational database of transcription factors. Nucleic Acids Res. 18:1749-1756. 13. Gillespie, F. P., L. Doros, J. Vitale, C. Blackwell, J. Gosselin, B. W. Snyder, and S. C. Wadsworth. 1993. Tissue-specific expression of human CD4 in transgenic mice. Mol. Cell. Biol. 13:2952-2958. 14. Gorman, S. D., B. Tourvieille, and J. R. Parnes. 1987. Structure of the mouse gene encoding CD4 and an unusual transcript in brain. Proc. Natl. Acad. Sci. USA 84:7644-7648. 14a.Hanna, Z., P. Poussier, and P. Jolicoeur. Unpublished data. 15. Hoxie, J. A., J. D. Alpers, J. L. Rackowski, K. Huebner, B. S. Haggarty, A. J. Cedarbaum, and J. C. Reed. 1986. Alterations in T4 (CD4) protein and mRNA synthesis in cells infected with HIV. Science 234:1123-1127. 16. Killeen, N., S. Sawada, and D. R. Littman. 1993. Regulated expression of human CD4 rescues helper T cell development in mice lacking expression of endogenous CD4. EMBO J. 12:15471553. 17. Klatzmann, D., E. Champagne, S. Chamaret, J. Gruest, D. Guetard, T. Hercend, J. C. Gluckman, and L. Montagnier. 1984. T-lymphocyte T4 molecule behaves as the receptor for human retrovirus LAV. Nature (London) 312:767-768. 18. Leung, S., S. McCracken, J. Ghysdael, and N. G. Miyamoto. 1993. Requirement of an ETS-binding element for transcription of the human Ick type I promoter. Oncogene 8:989-997. 19. Littman, D. R. 1987. The structure of the CD4 and CD8 genes. Annu. Rev. Immunol. 5:561-584. 20. Littman, D. R., and S. N. Gettner. 1987. Unusual intron in the immunoglobulin domain of the newly isolated murine CD4 (L3T4) gene. Nature (London) 325:453-455. 21. Maddon, P. J., D. R Littman, M. Godfrey, D. E. Madden, L. Chess, and R. Axel. 1985. The isolation and nucleotide sequence of a cDNA encoding the T cell surface protein T4: a new member of the immunoglobulin gene family. Cell 42:93-104. 22. Maddon, P. J., S. M. Molineaux, D. E. Maddon, K. A. Zimmerman, M. Godfrey, F. W. Alt, L. Chess, and R. Axel. 1987. Structure and expression of the human and mouse T4 genes. Proc. Natl. Acad. Sci. USA 84:9155-9159. 23. Maniatis, T., E. F. Fritsch, and J. Sambrook. 1982. Molecular cloning: a laboratory manual. Cold Spring Harbor Laboratory, Cold Spring Harbor, N.Y. 24. Means, A. L., and P. J. Farnham. 1990. Transcription initiation from the dihydrofolate reductase promoter is positioned by HIP1 binding at the initiation site. Mol. Cell. Biol. 10:653-661. 25. Melton, D. A., P. A. Kreig, M. R. Rebagliati, T. Maniatis, K. Zinn, and M. R. Green. 1984. Efficient in vitro synthesis of biologically active RNA and RNA hybridization probes from plasmids containing a bacteriophage SP6 promoter. Nucleic Acids Res. 12:7035-7056. 26. Mitchell, P. J., and R. Tjian. 1989. Transcriptional regulation in mammalian cells by sequence-specific DNA binding proteins. Science 245:371-378. 27. Paquette, Y., L. Doyon, A. Laperriere, Z. Hanna, J. Ball, R. P. Sekaly, and P. Jolicoeur. 1992. A viral long terminal repeat expressed in CD4+ CD8+ precursors is downregulated in mature peripheral CD4- CD8+ or CD4+ CD8- T cells. Mol. Cell. Biol. 12:3522-3530. 28. Poirier, Y., C. Kozak, and P. Jolicoeur. 1988. Identification of a
1094
HANNA ET AL.
helper provirus integration site in Abelson murine leukemia virus-induced lymphoma DNA. J. Virol. 62:39853992. Saizawa, K., J. Rojo, and D. Janeway. 1987. Evidence for physical association of CD4 and the CD3:a:p T cell receptor. Nature (London) 328:260. Salmon, P., A. Giovane, B. Wasylyk, and D. Klatzmann. 1993. Characterization of the human CD4 gene promoter: transcription from the CD4 gene core promoter is tissue-specific and is activated by Ets proteins. Proc. Natl. Acad. Sci. USA 90:77397743. Sands, J. F., and J. Nikolic-Zugic. 1992. T cell-specific proteinDNA interactions occurring at the CD4 locus: identification of possible transcriptional control elements of the murine CD4 gene. Int. Immunol. 4:1183-1194. Sanger, F., S. Nicklen, and A. R. Coulson. 1977. DNA sequencing with chain-terminating inhibitors. Proc. Natl. Acad. Sci. USA 74:5463-5467. Sawada, S., and D. R. Littman. 1991. Identification and characterization of a T-cell-specific enhancer adjacent to the murine CD4 gene. Mol. Cell. Biol. 11:5506-5515. common
29. 30.
31.
32. 33.
MOL. CELL. BIOL. 34. Sin, G., A. L. Wurster, J. S. Lipsick, and S. M. Hedrick. 1992. Expression of the CD4 gene requires a myb transcription factor. Mol. Cell. Biol. 12:1592-1604. 35. Smale, S. T., M. C. Schmidt, A. J. Berk, and D. Baltimore. 1990. Transcriptional activation by Spl as directed through TATA or initiator: specific requirement for mammalian transcription factor IID. Proc. Natl. Acad. Sci. USA 87:4509-4513. 36. Southern, E. M. 1975. Detection of sequences among DNA fragments separated by gel electrophoresis. J. Mol. Biol. 38: 503-517. 37. Veillette, A., M. A. Bookman, E. M. Horak, and J. B. Bolen. 1988. The CD4 and CD8 cell surface antigens are associated with the internal membrane tyrosine-protein kinase p56 lck cell. Cell 44:301-308. 38. Winoto, A., and D. Baltimore. 1989. Alpha beta lineage-specific expression of the alpha T cell receptor gene by nearby silencers. Cell 59:649-655. 39. Wood, G. S., N. L. Warner, and R. A. Warnke. 1983. Anti-Leu3/T4 antibodies react with cells of monocyte/macrophage and Langerhans lineage. J. Immunol. 131:212-216.
